Integrated circuits are often designed to incorporate scan test circuitry that facilitates testing for various internal fault conditions. Such scan test circuitry typically comprises scan chains, which are chains of flip-flops that are used to form serial shift registers for applying test patterns at inputs to combinational logic of the integrated circuit and for reading out the corresponding results. A given one of the flip-flops of the scan chain may be viewed as an example of what is more generally referred to herein as a “scan cell.”
In one exemplary arrangement, an integrated circuit with scan test circuitry may have a scan shift mode of operation and a functional mode of operation. A flag may be used to indicate whether the integrated circuit is in scan shift mode or functional mode. In the scan shift mode, the flip-flops of the scan chain are configured as a serial shift register. A test pattern is then shifted into the serial shift register formed by the flip-flops of the scan chain. Once the desired test pattern has been shifted in, the scan shift mode is disabled and the integrated circuit is placed in its functional mode. Internal combinational logic results occurring during this functional mode of operation are then captured by the chain of scan flip-flops. The integrated circuit is then once again placed in its scan shift mode of operation, in order to allow the captured combinational logic results to be shifted out of the serial shift register formed by the scan flip-flops, as a new test pattern is being scanned in. This process is repeated until all desired test patterns have been applied to the integrated circuit.
As integrated circuits have become increasingly complex, scan compression techniques have been developed which reduce the number of test patterns that need to be applied when testing a given integrated circuit, and therefore also reduce the required test time. However, the use of high levels of scan compression can adversely impact diagnostic resolution, that is, the ability to attribute a particular failure to an exact fault or set of faults within the combinational logic. As a result, when using scan compression, a tradeoff exists between compression level and diagnostic resolution. Additional details regarding compressed scan testing are disclosed in U.S. Pat. No. 7,831,876, entitled “Testing a Circuit with Compressed Scan Subsets,” which is commonly assigned herewith and incorporated by reference herein.
A significant problem that can arise in both compressed and noncompressed scan testing relates to the power consumption of the integrated circuit during scan testing. It is important that the test patterns generated and applied to the integrated circuit by a tester do not cause the integrated circuit to exceed its maximum power. A typical conventional approach to controlling scan test power consumption is implemented in software and performed during test pattern generation. Such an approach generally involves identifying test patterns that violate peak power limitations and then modifying those patterns before they are applied to the integrated circuit. An example of an approach of this type is described in R. Sankaralingam et al., “Controlling peak power during scan testing,” Proc. VLSI Test Symposium, 2002. However, this approach does not provide a sufficient level of controllability during the scan test itself. For example, it does not allow dynamic adjustment of the power consumption associated with particular applied scan test signals during scan testing.